The idea of nudging people towards positive behavior change has recently gained interest from academics, individuals, and even governments. Nudging can be explained in terms of social norming effects where we adjust our behavior to align with our peers or to what is perceived as a normative measure. Social-norm influencing can be communicated as feedback or messages. A descriptive norm message may be delivered, for example, as a comparative measure of how much energy you and others are using, while an injunctive norm message indicates whether the usage level is good or bad. A combination of these norming messages delivered over online social networks (OSNs) can create an effective environment in which nudging towards desirable behaviors can take place.
Much research has been carried out in facilitating the nudging of people towards positive changes in eating, fitness, and health through persuasive technology. So, what exactly is a persuasive technology? B.J. Fogg coined the term in his book Persuasive Technology and described it as “Technology that is designed to change attitudes or behaviors of the users through persuasion and social influence, but not through coercion.”
Despite recent interest in designing technologies to support behavior change, the process is not well understood by HCI researchers and professionals. Many interventions that use supporting technologies are poorly designed with little heed of the users’ needs and desires. For example, why implement an application that is designed merely for recording step counts on a complex smartphone when you can pick up an easier-to-use digital pedometer for a fraction of the price? The problem with this all-too-common approach is the layers of complexity introduced for such a simple task. Behavioral interventions should be designed with the user at the center of the process and be based on a full usability evaluation of currently available technologies.
At Lincoln Social Computing (LiSC) Research Centre, we see great potential in leveraging the engaging power of casual games and social applications to raise awareness about many individual and societal issues including, for instance, health, environment, and social responsibility.
Both case studies presented here were successful in bringing about positive behavior changes.
Case Study 1 – Energy in the Home
Our first completed study was based on reducing energy consumption in the home with the selected persuasive technology, a Wattson home energy monitor. Live energy data from the Wattson was socially displayed in the “Wattsup” Facebook application (see Figure 1).
With sustainability and energy security a global concern, our Wattsup study aimed to develop an understanding of a householder’s perceptions on energy usage in the home. We employed participatory design methodologies through focus groups to design Wattsup. A Wattson energy monitor coupled to Wattsup was trial-run in eight households and provided a social context for comparing their energy usage. Anecdotally, several householders were so elated upon having the monitor installed that they took to bounding around the house in front of the researcher turning everything off in an effort to get the display as close to zero as possible. If only this zealous behavior continued we would have no difficulty in changing behavior and reducing consumption. But it’s never that easy!
At the time of the study, home energy monitors where fairly rudimentary devices with the majority offering feedback solely through a static LCD display with no sharing or social context. Now a number of sophisticated monitoring devices, such as the Current Cost Bridge or AlertMe monitor, can move energy data online for use in interactive third-party consumer applications. Moving data into the “cloud” offers the opportunity for creative exploitation of online social interaction, providing more value than the monitor’s physical display alone.
Through focus groups we discovered potential users of Wattsup may prefer to see concrete metaphors of energy measurements to present their usage and to supplement raw energy units. The metaphoric value of visualizations in a persuasive technology is important, and ideally, they are designed by the users themselves (see Figure 2).
The Facebook application platform introduced a social context for participants’ energy usage, facilitating banter, competition, and engagement between friends to reduce energy consumption. Wattsup featured a rankings interface and a Facebook comments board, allowing friends to view and comment upon one another’s energy usage (see Figure 3). Banter included teasing such as “How come you are at the top, cut down drastically on your cups of tea?” And gloating “Good to see I’m higher in the table then you [anon] ha ha,” as well as provocations such as “Energy vampire…you clearly are!”
One feature of the engagement observed in the study was the participants’ willingness to disclose information such as “Left my main PC on the last two nights; made a massive difference to my scores.” And disbelief about energy usage: “Gone down in the rankings? I’m in Spain:S.” Social interaction was clearly evident in Wattsup, suggesting that the social features available to them through Facebook were a catalyst in helping reduce consumption.
The study ran in two conditions, one with Facebook’s social features—such as rankings, sending messages, and commenting—enabled, and the other with no social features. This design allowed us to test our hypothesis that energy data delivered in a social context among a group of friends can motivate reductions in consumption. Our results indicated seven of the eight households reduced their energy consumption while in the social condition with a total of 130 kilowatt hours saved.
Case Study 2 – Increasing Physical Activity at Work
Our second study was designed to target an increase in physical activity at work using a simple digital pedometer coupled with the Step Matron Facebook application.
Modern lifestyles are becoming increasingly sedentary. A U.S. governmental report states 60 percent of Americans are not physically active on a regular basis and includes 25 percent who are not active at all. Similarly, a UK governmental report states only 11.6 percent of adults are classed as physically active by taking part in moderate exercise five times a week or more. These changes in physical activity levels can be largely attributed to developments in technology. For example, we now choose to drive to the shop, take an elevator instead of stairs, and sit for many hours in front of a computer screen for work, leisure, and entertainment.
Our Step Matron study was comprised of the design and in-the-wild evaluation of a Facebook application for providing social and competitive context for daily pedometer readings, in order to motivate physical activity in the working environment. Step Matron aimed to determine whether social interactions between users via the application more successfully motivated physical activity than simply recording daily step counts with no social interaction. Nurses in a UK hospital were recruited and used the application across two conditions over the course of the study. In the socially enabled condition, participants could view each other’s step data and make comparisons and comments. In the non-social condition, participants could only view their own personal step data.
We paid careful attention, not only to the design of the Facebook application, but to the selection of a suitable pedometer based on HCI usability methodologies. The Silva pedometer was selected after a comparative usability evaluation on three commercially available digital pedometers. The pedometer with the lowest number of recorded minor-to-moderate usability problems was selected to ensure the study would run without any technical issues from the participants’ perspective.
The nurses self-reported their step count data as a task in the Step Matron Facebook application, which offered them the ability to compare their step data with their fellow nurses and to post comments on their peers’ activity. Each time a step count was entered, it was automatically posted to their Facebook profile news feed. This approach enabled all of the nurse’s friends on Facebook to view and post comments on the participants’ posted step counts.
A rankings interface displayed the total step count for each participant with a public comments board available for posting messages viewable by all, as shown in Figure 4. The rankings table provided the competitive and social attributes of Step Matron, as most of the nurses engaged with one other by commenting upon others’ step activity.
A significant increase in step activity was observed in the socially enabled condition with nine of the ten nurses increasing their step activity. Our findings also suggest that simple mobile devices can function as an inexpensive, accessible, and powerful trigger towards this behavior change without necessitating the use of overly complex and expensive mobile applications or devices.
Our findings in both case studies presented highlight the potential of social media as a means for generating positive behavior change. We are currently expanding our domestic energy research by investigating innovative ways to present energy feedback. Additionally, we have a research vein into organizational energy usage by utilizing expansive smart meter networks across higher education and local authority infrastructures. More information on the studies presented here as well as currently ongoing work can be viewed on our LiSC website at http://lisc.lincoln.ac.uk